With the aim of identifying a surrogate marker for the neurotoxic effects of methylmercury using a peripheral blood sample, the sensitivity of microtubules in circulating blood cells to depolymerization by methylmercury was compared. Methylmercuric chloride was added to samples of human venous blood or to isolated platelets and lymphocytes (human or rabbit) suspended in RPMI medium plus 10% fetal calf serum. After 1 h, microtubular networks were visualized by immunolabeling with antibody specific for tubulin. The percentage of platelets without visible, intact microtubules and the percentage of viable, unactivated lymphocytes without microtubules visibly radiating from the centriolar region through the cytoplasm were counted. A concentration-dependent loss of microtubules was observed in both cell types. Loss of microtubules was more easily quantitated and was observed at significantly lower concentrations in platelets compared to lymphocytes. The IC50 (concentration of methylmercuric chloride resulting in dissolution of microtubules in 50% of the cells) was 3.1 microM in platelets and 7.4 microM in lymphocytes in samples exposed in culture medium without erythrocytes. When methylmercury was added to whole blood for 1 h, the IC50 increased to 182 microM in platelets and >700 microM in lymphocytes, consistent with the known sequestration of methylmercury in erythrocytes. With longer durations of exposure, much lower concentrations of methylmercury were effective in both cell types. When rabbit lymphocytes and platelets were exposed to methylmercury under culture conditions, IC50s in platelets/lymphocytes were 2.5 microM/4.8 microM after 1 h of exposure, 0.77 microM/1.12 microM after 20 h, and 0.51 microM/0.63 microM after 70 h. The results of this study suggest that platelets may be more suitable than lymphocytes as a cell type in which to monitor in vivo effects of methylmercury on microtubules.